Yankees Outperforming Expectations

Sabermetrician Mitchel Lichtman contributed (along with Tom Tango and Andrew Dolphin) to one of the most data-intensive baseball books out there. The book, aptly named The Book: Playing the Percentages in Baseball, studies various baseball strategies such as when to employ a sacrifice bunt, what an optimal lineup looks like, and when to summon your best relief pitcher. MGL (his blogger name) started up a blog around the time of the playoffs last year and had some really interesting analysis to share. In preparation for the upcoming season, he took a look back at the past few years to see which teams overachieved and underachieved. The Yankees were extremely unique in his study.

MGL built a model to predict every team’s win-loss record over the past 5 years. The model incorporated the run scoring environment, starting pitchers, and relief pitcher and pinch hitter usage. He updated preseason projections after each month to include the current season’s performance. For instance, if a pitcher was expected to be around a 4 ERA-guy and then tanked in April and was hurt then his projection for the rest of the season would reflect a much lesser quality of pitcher. Additionally, MGL retrieved the closing line from Pinnacle Sports Book to also use in the analysis. A comparison was made between the Vegas line, MGL’s model, and actual team records.

The Yankees were an extreme outlier. The inputs in the model included a 98 RA9- (actual 99 FIP-) and a .30 expected marginal batting runs per game (actual was .45). They had a win percentage .039 points (equivalent to 6.3 wins per season) higher than what MGL’s model predicted. This was higher than every team in the MLB (beat the Phillies by .001). The actual record was also higher than Vegas’ win percentage projection by .019 points. Possible explanations include having the greatest closer of all-time in Mariano Rivera, having a “player’s manager” in Joe Girardi, and focusing on acquiring good-character guys to foster team chemistry.

Yankees’ fans love to talk about the mystique and aura that surrounds the team. Some thought that magic would evaporate as the team moved across the street into a new stadium. However, the past five years (’09-’13) indicate otherwise during the regular season. Teams tend to regress to the mean year-over-year due to the plexiglass principle posited by Bill James. Maybe this is pure luck as it is only a five-year sample. Maybe there is something that Vegas or MGL’s model doesn’t capture when evaluating the Yankees. The Yankees have the best win-percentage (.585) over this five-year stretch and it might just be difficult to project teams sitting on the upper range of the win curve. In any case, it will be interesting to see if the Yankees’ good fortunes continue during this transition phase away from the Core Four.